Volker Lendecke [Thu, 9 Jun 2016 19:27:09 +0000 (21:27 +0200)]
libreplace: Add a closefrom() implementation
There is closefrom in some BSDs, but Linux ships this only as part
of libbsd. Add a new implementation of it in libreplace. The one in
libbsd of jessie and upstream differ and it has for example optimizations
for FreeBSD, but it gets some of the array calculations slightly wrong
from my point of view. If you want those, use libbsd. This replacement
is optimized on Linux only looking at /proc/self/fd/, everything else
would do the OPEN_MAX brute force fallback.
Michael Adam [Wed, 8 Jun 2016 10:32:04 +0000 (12:32 +0200)]
ctdb: use properly configured ctdb in debug-hung-script.sh
Signed-off-by: Michael Adam <obnox@samba.org> Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Sat Jun 11 10:23:03 CEST 2016 on sn-devel-144
When we transitioned from autotools to waf we dropped installing the
findsmb script. However we create and install the manpage for it so
install it in the system again.
Signed-off-by: Andreas Schneider <asn@samba.org> Reviewed-by: Guenther Deschner <gd@samba.org>
Autobuild-User(master): Andreas Schneider <asn@cryptomilk.org>
Autobuild-Date(master): Thu Jun 9 19:10:15 CEST 2016 on sn-devel-144
Signed-off-by: Jeremy Allison <jra@samba.org> Reviewed-by: Uri Simchoni <uri@samba.org>
Autobuild-User(master): Uri Simchoni <uri@samba.org>
Autobuild-Date(master): Thu Jun 9 13:18:56 CEST 2016 on sn-devel-144
Amitay Isaacs [Wed, 8 Jun 2016 05:04:52 +0000 (15:04 +1000)]
ctdb-daemon: Reset push_started flag once DB_PUSH_CONFIRM is done
Once DB_PUSH_START is processed as part of recovery, push_started
flag tracks if there are multiple attempts to send DB_PUSH_START.
In DB_PUSH_CONFIRM, once the record count is confirmed, all information
related to DB_PUSH should be reset. However, The push_started flag was
not reset when the push_state was reset.
Signed-off-by: Amitay Isaacs <amitay@gmail.com> Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Amitay Isaacs <amitay@samba.org>
Autobuild-Date(master): Wed Jun 8 14:31:52 CEST 2016 on sn-devel-144
In do_recovery, after the recovery and takeover is complete, recoverd
event is triggered. When the parallel database recovery was separated,
ctdb_recovery_helper implemented sending END_RECOVERY control which
causes recoverd event to be triggered. So when there is parallel database
recovery, recoverd event is triggered twice.
Instead move the call to run_recovered_eventscript() explicitly in
the serial recovery code path. This avoids the duplication trigger of
recoverd event.
Signed-off-by: Amitay Isaacs <amitay@gmail.com> Reviewed-by: Martin Schwenke <martin@meltin.net>
Martin Schwenke [Mon, 6 Jun 2016 03:56:55 +0000 (13:56 +1000)]
ctdb-scripts: Move NFS callout-related code to functions file
A second NFS eventscript may be required, so make this code available
to it.
The initialisation code can't be evaluated in the functions file
because service_state_dir isn't yet setup, so put it in a function and
call it with other initialisation code.
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Andrew Bartlett [Thu, 2 Jun 2016 21:53:29 +0000 (09:53 +1200)]
build: Address may be used uninitialized in this function on Ubuntu 10.04
This is not found by modern compilers, but prevents the -Werror -O3 build on Ubuntu 10.04
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Jun 8 08:48:57 CEST 2016 on sn-devel-144
Andrew Bartlett [Thu, 2 Jun 2016 21:23:12 +0000 (09:23 +1200)]
build: Try to work around strict aliasing rules on Ubuntu 10.04
We get cc1: warnings being treated as errors
../lib/util/util_net.c: In function get_socket_port:
../lib/util/util_net.c:921: error: dereferencing pointer sa.106 does break strict-aliasing rules
../lib/util/util_net.c:921: note: initialized from here
../lib/util/util_net.c:925: error: dereferencing pointer sa.107 does break strict-aliasing rules
../lib/util/util_net.c:925: note: initialized from here
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Michael Adam <obnox@samba.org>
Amitay Isaacs [Mon, 6 Jun 2016 05:04:08 +0000 (15:04 +1000)]
ctdb-cluster-mutex: Fix #endif decoration
Signed-off-by: Amitay Isaacs <amitay@gmail.com> Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Martin Schwenke <martins@samba.org>
Autobuild-Date(master): Wed Jun 8 04:52:18 CEST 2016 on sn-devel-144
Martin Schwenke [Tue, 1 Mar 2016 01:32:48 +0000 (12:32 +1100)]
ctdb-tool: Drop support for "ctdb setreclock" command
The recovery lock can not be reliably updated at run-time. If it
fails to update on some nodes then split-brain protection is gone and
there is no reasonable way to repair the situation. CTDB will have to
be restarted on all nodes. So, if this feature is being used to avoid
scheduling an outage then an outage will have to be scheduled just in
case!
To update the recovery lock, shut down CTDB on all nodes, reconfigure
the recovery lock and start CTDB again.
Those that *really* want to be able to change the recovery lock at
run-time can still do so. Set CTDB_RECOVERY_LOCK to point to a script
and this script can then be modified at run-time. However, please
don't report bugs if bad things happen...
Signed-off-by: Martin Schwenke <martin@meltin.net> Reviewed-by: Amitay Isaacs <amitay@gmail.com>
Volker Lendecke [Tue, 7 Jun 2016 08:07:21 +0000 (10:07 +0200)]
libsmb: Fix two CIDs for NULL dereference
This whole area is a known-to-be-broken mess, but this patch should fix
the immediate crash
Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Michael Adam <obnox@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Tue Jun 7 18:31:30 CEST 2016 on sn-devel-144
Andrew Bartlett [Tue, 7 Jun 2016 04:41:15 +0000 (16:41 +1200)]
repl: Avoid use-after-free when working with the working_schema
The original schema must live as long as the working_schema
as the working_schema starts as a shallow-copy of schema.
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=11953
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Tue Jun 7 14:33:39 CEST 2016 on sn-devel-144
Andrew Bartlett [Wed, 25 May 2016 02:49:31 +0000 (14:49 +1200)]
dsdb: Fix rename and RDN handling for replPropertyMetaData
This matches Windows 2012R2, which both has the RDN not sorted last and has it updated with the local
invocation_id and a local version.
The RDN attribute, unlike name, is not replicated over DRS, so the impact for interopability extends only to
the incorrect RDN values that we were finding with dbcheck (values that did not match the name values).
Finally, we always force the RDN to match the name attribute, which avoids issues
in dbcheck where these diverge. As such, we can finally remove dbcheck as a
flapping test, last re-added in e4bab3a8282d263eb2391bc7e8a6fd64ae068935
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Andrew Bartlett [Tue, 7 Jun 2016 01:51:09 +0000 (13:51 +1200)]
selftest: Run the krb5.kdc test on a more selective basis
The previous tests would take 20mins, the new set of tests take around 7 mins and still cover
the important combinations, given that it is the same KDC code in each environment
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Andrew Bartlett [Tue, 19 Apr 2016 05:28:59 +0000 (17:28 +1200)]
selftest: These replication tests are now OK after we fixed all the replication bugs
We have run this test a number of times, and these tests do not flap
in those tests
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Tue Jun 7 01:45:08 CEST 2016 on sn-devel-144
Andrew Bartlett [Tue, 19 Apr 2016 01:37:19 +0000 (13:37 +1200)]
selftest: Add tests to show that we can not create duplicate schema entries
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Mon Jun 6 20:32:54 CEST 2016 on sn-devel-144
Andrew Bartlett [Mon, 18 Apr 2016 02:59:21 +0000 (14:59 +1200)]
Remove the try/catch from urgent_replication.py
This meant that for ages, the duplicate OID was unnoticed, and when the syntax
was corrected recently, this caused the test to run, and so cause trouble
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz
Andrew Bartlett [Wed, 9 Dec 2015 01:00:10 +0000 (14:00 +1300)]
selftest: initial version of new repl_move test
This tests complex rename and modify combinations in a way that
demonstrated a number of replication failures, due to incorrect
handling in Samba when the parent of the record changes.
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz
The previous behaviour of this code was to overwrite the req->callback of the original, failed request.
This is a problem for many reasons - including that ldb_module_done() may already have been
called on that pointer.
The correct pattern is to create a new request, and to call ldb_module_done() on the parent
request (the one in ar->req) not in this one, in the error case.
We use the passed in new callback either as the callback to call on success, or
as the callback to the ADD request. We overwrite it with replmd_op_name_modify_callback
in the rename remote case, as before, but no longer modify req->callback as
this will not be used again.
This is less tricky and a little simpler to follow, as we also remove the
now unused handling for RENAME, which is in a separate routine now
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Andrew Bartlett [Mon, 21 Mar 2016 00:32:55 +0000 (13:32 +1300)]
dsdb: Split rename case out of replmd_op_possible_conflict_callback
This avoids running this code path, originally written for the add case, in
a semi-async manner in the rename case, which caused both bugs and complexity.
This does create a deal of duplicated code, but it is easier to follow because
there are no longer special cases for ADD and RENAME in the "common" code and
the behaviour of ldb_module_done() and the callbacks is well defined and expected
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz
Andrew Bartlett [Wed, 9 Dec 2015 04:05:56 +0000 (17:05 +1300)]
repl: Enforce that we have parent objects for all replicated objects
The creating of replicated objects without their parent object allows database corruption as they can end up under
the wrong object. We need to re-try the replication with the DRSUAPI_DRS_GET_ANC flag
set to get the objects in tree order.
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz>
Peter Somogyi [Wed, 1 Jun 2016 16:30:08 +0000 (18:30 +0200)]
Add yet another error code when forking an smbd and ctdb is not there. We can see NT_STATUS_CONNECTION_REFUSED in the logs upon such a rare case.
Signed-off-by: Peter Somogyi <psomogyi@hu.ibm.com> Reviewed-by: Volker Lendecke <vl@samba.org> Reviewed-by: Martin Schwenke <martin@meltin.net>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Jun 6 16:10:19 CEST 2016 on sn-devel-144
Commit 670db6ac1d678babd25dd82c4467c0f094cfabc5 split tevent-util public
library to create tevent-unix-util public library for standalone ctdb
use. This created a public library dependency between samba and ctdb
for packaging.
Bundle tevent_unix.c in public library tevent-util as before. However,
to avoid the dependencies for packaging, standalone ctdb build will
build tevent-util as a private library with only tevent_unix.c
This simplifies any new subsystems (or libraries) which need tevent-util
and are linked in both samba and ctdb.
Andrew Bartlett [Thu, 26 May 2016 22:22:15 +0000 (10:22 +1200)]
dsdb: Cache our local invocation_id at the start of each request
This avoids fetching it over and over again
Signed-off-by: Andrew Bartlett <abartlet@samba.org> Reviewed-by: Garming Sam <garming@catalyst.net.nz
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Mon Jun 6 12:25:14 CEST 2016 on sn-devel-144
projects / thirdparty / samba.git / log
